Acupuncture in cancer pain

Systematic reviews of the evidence on cancer pain include a Cochrane review published in 2015 and at least 11 other systematic reviews since that date. The scope of the acupuncture intervention and type of pain varied across the reviews. Some reviews also included study designs in addition to RCTs. Therefore, it is difficult to compare results. The systematic reviews are described in table 1.  Collectively, the evidence on reducing cancer-related pain is not conclusive.

The Society for Integrative Oncology (SIO) has produced evidence-based guidelines on use of integrative therapies during and after breast cancer treatment (Greenlee 2017). These have been endorsed by the American Society of Clinical Oncology (Lyman 2018). The SIO guidelines are based on a systematic review of RCTs published up to 2015 and state that, although the 5 trials located had small sample sizes and mixed findings, acupuncture can be considered for pain associated with aromatase inhibitor associated musculoskeletal symptoms.

Description of included studies

Acupuncture

A Cochrane review of acupuncture for cancer pain was updated in 2015 (Paley 2015) and confirmed as stable in 2020. A total of 5 RCTs including 285 participants were identified. One study included people with chronic peripheral or central neuropathic pain related to cancer. Auricular acupuncture was compared with auricular acupuncture at 'placebo' points and with non-invasive vaccaria ear seeds (Semen vaccariae) attached at 'placebo' points. Participants in the true acupuncture group reported lower pain scores at two months than those in the other groups. The study was assessed as high quality but there was a question over whether blinding was maintained, which may have biased the results. One trial compared acupuncture with medication for unspecified cancer and another trial compared acupuncture with medication and with point-injection for stomach carcinoma. Both reported positive results favouring acupuncture but the reliability of the results is uncertain due to methodological limitations, small sample sizes, poor reporting and inadequate analysis.

The two further RCTs included when the review was updated in 2015 compared electroacupuncture with sham acupuncture in pancreatic cancer pain and gynaecological cancer-related pain respectively. One reported positive results but blinding was unclear while the second did not find a difference but may not have been of sufficient size to do so. Meta-analysis was not carried out due to variations in the methodologies, cancer populations and techniques used and, overall, there was insufficient evidence for firm conclusions.

Reviews published subsequently have had a different focus or scope from the Cochrane review. This is reflected in number of RCTs included in each review which ranges from 1 to 41. There is also little overlap in the trials included in the various reviews. Five reviews included studies of a wider range of pain experienced by cancer patients (i.e. not only pain directly related to cancer but also pain related to treatments or surgery) (Hu 2016; Chiu 2017; Behzadmir 2020; He 2020; Dai 2021). Four reviews include acupressure as well as acupuncture (Behzadmir 2020; He 2020; Lau 2016; Yang 2020) while one covers various therapies (Dai 2021). Two reviews focus on pain as an adverse effect of cancer treatment (Pan 2018; Liu 2021)while three focus on pain in cancer palliative care (Dai 2021; Lau 2016; Yang 2021).

Those that are most up-to-date based on when searches for trials were carried out are the reviews by Dai (2021), Dong (2021), Yang (2020) and Yang (2021). All four reviews included trials published in English or Chinese and are described below:

Dai’s review (2021) covered various types of pain in cancer patients receiving palliative care. It also covered various acupuncture and related treatments (e.g. TENS) and studies other than RCTs. The 41 RCTs that were included tested the effects of combination treatment against analgesics alone. Only 2 RCTs had been included in the Cochrane review. Combined acupuncture and analgesics resulted in greater reduction in pain scores than analgesics alone (weighted mean difference [WMD]: 1.33 [0.85–1.82], p<0.001). The quality of the evidence was, however, rated as low.

The review by Dong (2021) focused on wrist acupuncture. Of the 8 RCTs, only 1 had been included in the Cochrane review . Acupuncture combined with analgesics was found to be more effective than analgesic alone (RR for pain relief = 1.31, 95% CI: 1.15-1.49, p<0.01]. No trials were blinded and the risk of bias was judged high.

Yang’s review (2020) focused on auricular therapy using acupuncture or acupressure. Two of the 9 RCTS had been included in the Cochrane review. The authors reported that auricular therapy combined with drug therapy was more effective than drug therapy alone based on effective rate for pain relief (RR = 1.40; 95% CI 1.22, 1.60; 4 studies). They also reported reduced adverse effects (RR = 0.46; 95% CI 0.37, 0.58) and a difference between acupuncture and sham (SMD = -1.45; 95% CI -2.80, -0.09; 2 studies). The majority of studies were at unclear risk of bias.

Yang’s review (2021) covered pain in palliative cancer patients but only 1 of the 5 included studies was an RCT and the reliability of the conclusions indicating an effect is not clear.

Of the remaining 7 reviews, two focus specifically on pain caused by cancer treatments (Liu 2021; Pan 2018). A review focused on breast cancer hormone therapy-related side effects did not find any effect on pain based on four small trials (Pan 2018). Lui’s 2021 review did find a beneficial effect on arthralgia caused by aromatase inhibitors based on 7 RCTs (mean difference in pain severity −1.57, 95% CI [−2.46, −0.68]). Four of the studies were judged at high risk of bias on one of the criteria.

Group acupuncture

A non-inferiority trial compared group acupuncture whereby patients are treated in a group setting for single conditions using standardised or semi-standardised protocols (Berkovitz 2008) with individual acupuncture for cancer pain (Reed et al. 2020). Seventy-four participants were randomly allocated to twelve treatments. Randomisation and allocation concealment are not described and blinding was not possible.

The primary outcome was pain (interference and severity) as measured by the BPI-SF. Overall attrition was 30.7% and significantly more participants receiving individual acupuncture (80%) completed both pre-assessment and post-assessment than those in the group acupuncture (58%). Both arms showed statistically significant improvements across all symptoms before and after the intervention.

Group acupuncture was found to be noninferior to individual acupuncture for treating cancer pain and was superior in many health outcomes including cost. Further confirmation of these findings are needed.

Acupressure

No systematic reviews were located that focused specifically on acupressure.

Several trials were included in systematic reviews described above. Behzadmir’s review (2020) included one trial (Zick 2018, described below). He’s review (2020) included two trials of acupressure. These were published in Chinese so few details are readily available but both appear to be open-label trials comparing of acupressure with analgesic versus analgesic alone in bone pain (n=60) and malignant neuropathic pain (n=46) respectively. Both reported positive results but were at high risk of bias.

A third trial in He’s review, combined acupressure with acupuncture so  the effects of acupressure alone could not be determined. Yang’s review (2020) included a trial of auricular therapy (n=42) using press needles combined with analgesic versus analgesic alone. Positive results were again reported but risk of bias was judged unclear.

Three further RCTs focusing on acupressure were located: a trial of acupressure in leukaemia patients (Sharif 2017), a trial in breast cancer patients with symptom clusters that included pain (Yeh 2016), and an assessment of self-acupressure in cancer survivors (Zick 2018). Two further open-label studies were included in a systematic review described under Acupuncture above (He et al 2020) .

The RCT of acupressure for cancer pain in 100 hospitalised leukaemia patients  found no significant differences between 12 acupressure sessions added to treatment and standard treatment alone (Sharif 2017). A second small, pilot RCT (n=31) assessed the effects of ‘true’ versus sham acupressure for symptom clusters of pain, fatigue and sleep problems in patients with breast cancer (Yeh 2016). Between group differences were seen in pain and distress reflecting beneficial effects of acupressure at the end of the intervention but not at one month.

One further RCT (n=288) involved daily self-administered acupressure using a ‘relaxing’ and a ‘stimulating’ protocol compared with usual care for 6 weeks (Zick 2018). The participants had stage 0 to III breast cancer and had completed primary treatment at least 12 months previously and were suffering from fatigue. Pain was measured using a visual analogue scale and the Brief Pain Inventory. Post-hoc analysis showed that relaxing acupressure was associated with greater reductions in pain severity, and stimulating acupressure was associated with greater reductions in pain interference after treatment. These effects were not maintained at 10 weeks.

References